AWS Transcribe Processor

The AWS Transcribe Processor initiates transcription jobs (speech to text) on AWS Transcribe, given an S3 Bucket Object for input and output.

Processor (Adapter) Configuration Drop-Down List

Select the AWS Transcribe Processor from the drop-down list and hit Add.

Click on Add

Basic AWS Transcribe Processor Configuration Options

On the Basic tab, you can specify:

Transcription Job Name – unique ID of the transcription job
Source Audio S3 Object – the S3 bucket object location of the audio file to be transcribed. If the object is not already there, use an AWS S3 module to upload it before this step. The string Format should be as follows: bucketName/path/to/object.mp3
Transcription Output Bucket – the S3 bucket object location where the transcription will be placed after job completion. It can only be the bucket name with no sub-object folders.
Input Language – the language of spoken audio in the input file
Input Media Format – select the media format of the input file, your options are as follows: MP3, MP4, WAV, FLAC, OGG, AMR, WEBM or M4_A

AWS Transcribe Processor Basic Configuration Options

You can set the necessary credentials for the connection on the Credentials tab.

Region – specify the AWS Region to connect to
Use Credentials from – specifies the method of getting the credentials; note choosing the No Credentials option will only allow access to public buckets
Access Key ID – Access Key ID for making calls to Amazon services
Secret Key – Secret Access Key for making calls to Amazon services
Session Token – the STS session token for making calls to Amazon services. If you are not using STS, leave it blank.
AWS File Path – specifies the file path where the AWS profile configuration file is located
Profile Name – specifies the name of a local configuration profile

AWS Transcribe Processor Credentials Configuration Options